DEVELOPMENT OF AN APPROACH TO CLUSTERING OF DISTRICTS BASED ON THE KNIME MACHINE LEARNING TOOL
Abstract
The scientific work is devoted to the application of clustering methods for grouping homogeneous areas within larger territorial formations to organize reasonable state financing of territorial development. At the present time for the grouping of districts a number of statistical indicators are used which do not always reflect the real situation and do not allow to reasonably group heterogeneous districts within one area. Clustering for clustering purposes can become a tool of budget policy that allows to reasonably allocate funds based not only on social and demographic indicators, but also on a number of additional factors. Districts with similar indicators of socio-economic development may have different potential and need different financing strategies. To make informed long-term decisions the application of clustering method should be based on a number of statistical indicators over a long period of time. This paper presents an approach to the clustering of districts by the example of Ivanovo region. The regional center is singled out into a separate cluster, which is substantiated by significant differences from other settlements according to all indicators. Clustering is a method of machine learning without a teacher, that is, the data will be grouped without the analyst specifying the criteria of division
References
National Program «Digital Economy of the Russian Federation», approved by the Presidium of the Presidential Council for Strategic Development and National Projects, Protocol of 04.06.2019 N7. (in Russian).
Valinurova A.A. Theoretical substantiation of the use of clustering method in bank reservation. Ivecofin. 2021. N2(48). P. 17-22. (in Russian).
Valinurova A.A., Danilova S.V. A new approach to the formation of reserves in a commercial bank. Bulletin of the Ivanovo State University. Series: Economics. 2020. N4(46). P. 93-97. (in Russian).
Valinurova A.A., Smirnova E.M., Valinurov T.R., Bal-yasova E.D. Transformation of approaches to assessing the effectiveness of a bank branch in the digital economy. Bulletin of the Ivanovo State University. Series: Economics. 2020. N2(44). P. 90-97. (in Russian).
Valinurova A.A., Smirnova E.M., Ksenofontova O.L. Intelligent remote banking and its features. Modern high-tech technologies. Regional application. 2021. N2(66). P. 16-21. (in Russian).
Danilova S.V. Actual problems of business analytics and data mining in the digital economy. Materials of the scientific and practical Festival «Research activity in the classical University: traditions and innovations». 2020. P. 386-388. (in Russian).
Dubova S.E. Analysis of the potential for the spread of the phenomenon of digital money of central banks by credit organizations and institutions of the non-financial sector of the economy. Ivecofin. 2021. N3(49). P. 59-68. (in Russian).
Dugushkina N.V. Review of popular clustering methods in machine learning. Scienceasphere. 2020. N7. P. 112-118. (in Russian).
Kamaev V. A., Shcherbakov M. V., Al-Kataberi A.S. Analysis of socio-economic development of municipalities on the example of the Volgograd region using data mining models. Bulletin VSTU. Series «Actual problems of management, computer technology and informatics in technical systems". 2010. Vol. 6. N8. P. 103-106. (in Russian).
Krevsky M.I., Smetanev D.M. Application of machine learning and process mining to the tasks of clustering processes in the financial sector. Modern tools, methods and technologies of knowledge management. 2020. N3. P. 52-60. (in Russian).
Kuznetsov D.Yu., Troshina T.L. Cluster analysis and its application. Yaroslavl Pedagogical Bulletin. 2006. N4(49). P. 103-107. https://cyberleninka.ru/article/n/klasternyy-analiz-i-ego-primenenie/viewer. (in Russian).
Merzlikina G.S., Kuzmina E.V. Methodology for assessing the effectiveness of cluster activities based on a balanced scorecard in the digital economy. Ivecofin. 2020. N2(44). P. 14-23. (in Russian).
The Ministry of Digital Development, Communications and Mass Communications of Russia will coordinate the development of new production and communication technologies. https://digital.gov.ru/ru/events/41196 (in Russian).
Mirkin B.G. Cluster analysis methods for decision support. Moscow: HSE. 2011. 88 p. (in Russian).
The official website of the KNIME platform. https://www.knime.org. (in Russian).
Tyurin A.G., Zuev I.O. Cluster analysis, clustering methods and algorithms. Bulletin of MSTU MIREA. 2014. N2(3). P. 86-97. (in Russian).
Usman M. Hierarchical clustering using Python and Scikit-learn. https://pythobyte.com/hierarchical-clustering-with-python-and-scikit-learn-768953fe. (in Russian).
Khomyakova A.A., Mizgirev L.S., Shergin V.V. The use of data mining methods in the processes of managing the investment attractiveness of the region. Ivecofin. 2020. N2(44). P. 14-23. (in Russian).
Gorban А., Kegl B., Wunsch D., Zinovyev A. Principal Manifolds for Data Visualization and Dimension Reduction. Berlin – Heidelberg – New York. 2007.
Kanungo T. An efficient k-means clustering algorithm: Analysis and implementation. Pattern Analysis and Machine Intelligence. 2002. Vol. 24. N 7. P. 881–892.
